(149ac) Optimization-Based State and Parameter Estimation for Distributed Parameter Pipeline Systems

As one of the most representative and cost-effective infrastructures for material transportation, pipelines have been widely utilized to transport hydrocarbon liquid commodities such as petroleum, crude oil, and diesel fuel. With the growing requirements for energy transportation, monitoring and controlling a pipeline system safely and efficiently are important [1]. However, in practice, it would be prohibitively expensive to install pressure and flow meters throughout entire pipeline systems. In addition, parameters (i.e., pipe friction) are often not accurately known due to surface roughness and fluid parameters, making it difficult to model accurately. Therefore, accurate state and parameter estimation based on limited measurements is crucial in describing complex dynamics and accounting for uncertainties in pipeline systems.

As a typical distributed parameter system, the flow dynamic behaviors depend on time and space. An infinite-dimensional transient hydraulic model is introduced to describe the complex flow dynamics within a liquid pipeline [2]. In this work, to develop an accurate estimator, we formulate and solve one optimization-based problem to realize an online estimation of states (i.e., pressure and flow velocity) and parameters (i.e., friction coefficient) based on the available plant or input information and measurement that are corrupted with bounded disturbances [3]. The proposed optimization framework can take the physical constraints into consideration. In particular, the Cayley-Tustin transform is utilized to convert the continuous infinite-dimensional system into a discrete one without spatial discretization or model order reduction [4]. To improve the calculation efficiency, a receding horizon implementation strategy is further proposed, similar to [5]. The effectiveness of the proposed designs is finally verified via case studies. Sensitivity studies are provided to show the robustness of the proposed designs.

References:

[1] K. Sundar and A. Zlotnik. “State and parameter estimation for natural gas pipeline networks using transient state data,” IEEE Transactions on Control Systems Technology, 27, no. 5 (2018): 2110-2124.

[2] Xie, J., Xu, X. and Dubljevic, S., 2019. Long range pipeline leak detection and localization using discrete observer and support vector machine. AIChE Journal, 65(7), p.e16532.

[3] A. Alessandri and M. Awawdeh. Moving-horizon estimation with guaranteed robustness for discrete-time linear systems and measurements subject to outliers, Automatica, 67, pp.85-93, 2016.

[4] Havu V., Malinen J., 2007. The Cayley transform as a time discretization scheme. Numerical Functional Analysis and Optimization 28 (7-8): 825-851.

[5] J. Jalving and V. M. Zavala. “An optimization-based state estimation framework for large-scale natural gas networks,” Industrial & Engineering Chemistry Research, 57, no. 17: 5966-5979, 2018.